North American Prairie Conference
Date of this Version
1989
Abstract
Landsat thematic mapper digital data was found to offer an excellent potential for regular monitoring of the tallgrass prairie ecosystem by providing estimates of aboveground biomass production. Data from seven channels of a May thematic mapper scene were analyzed individually and in various combinations using stepwise regression in Statistical Analysis System (SAS). These procedures were used to determine the most appropriate multiple regression equation for estimating production of 1) total live aboveground biomass, 2) grasses, 3) forbs, 4) previous years dead, and 5) current years dead. Regression equations were based on satellite-derived estimates relative to ground level biomass values for watersheds on Konza Prairie Research Natural Area under a variety of burning treatments. Results suggest that multiple channel equations were most appropriate for measuring production of forbs and total live aboveground biomass. Channel one (0.45 to 0.52 μm) and channel four (0.76 to 0.90 μm) were applicable to estimate production of grass and levels of previous years litter, respectively. However, none of the channels were accurate in predicting current years dead. Further plans involve using thematic mapper data to estimate aboveground biomass over an entire growing season on Konza and exploring the potential of using satellite data to monitor grassland production across the Great Plains.
Comments
Published in Prairie Pioneers: Ecology, History and Culture: Proceedings of the Eleventh North American Prairie Conference, August 7-11, 1988, Lincoln, Nebraska (Lincoln, NE 1989).